technology improving your meltshop performancefoundryinfo-india.org/images/pdf/3b1.pdf · the...

| 1 |

TRANSACTIONS OF 61st INDIAN FOUNDRY CONGRESS 2013

INTRODUCTION

Every casting (Fig.1) begins with molten metal and every

foundry wants to make a profit. To melt metal, it requires

energy and in the case of an induction furnace it means

kilowatts.

More efficiently one uses those kilowatts, lower is the

costs and the greater is the profit.

Fig. 1: Typical automotive castings.

There are “technologies” and practices out there designed

to help, and the basis of this paper is to give a better

understanding of what these are and how they can be

implemented in an iron foundry.

Once it is established about how much molten metal is

required per “batch” the frequency at which a “batch” is

required, in conjunction with the “utilisation” would

determine the kilowatts are required and the total

production would determine the quantity of power

supplies required.

The following are the stages one must go through to get a

ladle full of metal with a specific analysis at the required

temperature:

MELT CYCLE - is the period when maximum power is

continuously applied to the furnace and charge is added.

Technology Improving YourMeltshop Performance

NON-PRODUCTION CYCLE - is when no or reduced

power is being applied, such as when the initial charge is

being added, when slag is being removed, when a

temperature dip or analysis sample is being taken, waiting

for an analysis result and pouring the furnace empty.

PRODUCTION CYCLE - is the “melt cycle” and the

“non-productive cycle”.

UTILISATION - is the “melt cycle” divided by the

“production cycle” expressed as a percentage. If the

melting cycle is of 45 minutes and the “non-production”

cycle is of 45 minutes, then the “production cycle” is 90

minutes. The 45-minute “melt cycle” divided by the 90-

minute “production cycle” times 100 gives an “utilisation”

of 50%.

If it is a process that requires 10 tonnes of iron poured

into moulds per hour and the “production cycle” is such

that it can only achieve 50% “utilisation”, one will have

to buy a power supply capable of melting 20 tonnes per

hour.

CHARGE MATERIALS - for producing gray and

ductile iron, the charge consists of steel scrap, pig iron,

in-house returns and scrap.

CHARGE PREPARATION and CHARGING -

irrespective of whether the furnace is to be charged

manually or mechanically, the charge materials should

be weighed and these ought to fit into the furnace. If the

scrap sections are long and extend out of the top of the

furnace, these, will ultimately melt but will take time,

which will influence the “utilisation”. The initial charge

needs to be as quick as possible and of sufficient density

to allow maximum power. The initial charge needs to be

added to the furnace as quickly as possible. For optimum

performance, the density should not be less than 1,750

kg per cubic metre and the furnace must be filled to a

minimum of 30% and maximum of 60% of its rated

capacity.

Graham Cooper

| 2 |


SLAG REMOVAL - an essential but unpleasant job.

The slag produced during melting iron is usually viscous

and, if not, the addition of a pearlite-based coagulant will

create a viscous slag that can be raked or lifted off. For

any sized furnace removing the slag manually is possible,

but on larger furnaces it is unpleasant and time

consuming. The options for larger furnaces are “back tilt

slag removal” or a slag grab.

TEMPERATURE DIP AND SAMPLE ANALYSIS -

a temperature dip to establish the bath temperature and

a molten metal sample has to be taken to establish the

chemical composition.

CHEMISTRY ADJUSTMENT AND SUPER-

HEATING – as soon as the analysis result is known the

“trim” materials should be weighed and charged into the

furnace and full power applied to superheat to the pouring

temperature.

POURING CYCLE - the time taken to empty the

furnace.

To help one to achieve an optimum “production cycle”

one needs to understand fully the technologies and

practices that are available.

THE CHARGE - Every “melt cycle” starts with a charge

(Fig. 2). The charge materials must be weighed before

these are put into the furnace. When the charge is fully

molten it is easier to calculate any “trim” additions, if the

weight of the liquid iron is known.

If “standard” charges for various grades of iron are

developed, it is much easier to achieve the target analysis.

One will be able to use more lower cost materials, as one

will learn the addition and recovery rates for the various

materials used for element additions.

The quality of these materials and the charging sequence

are important. Rust on steel scrap can influence lining

life. Sand on in-house returns has to be melted and then

removed as slag. Purchased cast iron scrap may contain

unwanted elements and must be purchased with care. To

melt slag, energy required is twice as much as that for

iron.

Alloying materials also influence performance of the “melt

cycle”. As a base recarburiser, one can use a lower cost

petroleum coke (Fig. 3).

Fig. 3: Petroleum coke.

If this is for gray iron, materials with higher sulphur levels

can be used as final sulphur of 0.5 to 0.12%, enhances

the performance of ferro silicon-based inoculants. Silicon

addition should be made using 75% silicon ferrosilicon

as this is an exothermic reaction. An alternative source

of silicon is silicon carbide which is 66% Si, 33% C and

zero carbon and an important consideration if one is

making a ductile or compacted graphite iron.

At the end of the melt, it is often necessary to make small

adjustments to “trim” the final analysis.

The charging sequence will influence carbon pick-up. Pig

iron is high in carbon and the carbon is already in the

iron matrix. It should be charged late in the “melt cycle”.

In-house foundry returns and scrap are of known analysis

Fig. 2: Typical charge materials.

| 3 |


with the elements in the matrix so they can be added last.

Steel is low in carbon and carbon can enter the matrix.

The charging sequence should be steel scrap, recarburiser,

in-house returns and scrap, pig iron (if required) and ferro

alloys.

Any “trim” materials should be added after the slag has

been removed and the charge is being superheated.

The furnace should not be overfilled. Any molten metal

above the power coil at the end of the “melt cycle” reduces

the surface stirring and subsequent surface, when “trim”

materials are to be added. Movement caused by thermal

convection and carbon boil does not enhance element

recovery.

For any high production, a mechanised charging system

is essential. The initial step is to weigh the charge. A

crane scale can be used to weigh a charge as it is loaded

into a feeder. A furnace feeder can be mounted on a

track weighing station. A weigh hopper (Fig. 4) can be

positioned to discharge into a feeder.

Fig. 4: Feeder and weigh hopper.

The best option is to select a feeder that can hold a full

furnace charge. If it is a large furnace, then a feeder thatholds half the charge can be used but, in this case, a weighhopper also capable of holding half the charge is essentialto ensure no delays in the charging sequence that couldaffect the “utilisation”.

The size of the scrap is important to ensure the chargedoes not bridge. On average, each piece should not havea dimension greater than 33 % of the furnace diameterand no dimension should ever exceed 50% of the furnace

diameter. The feed rate of the system must be able to

deliver the full charge into the furnace within 65 to 70%

of the actual “melt cycle”.

The equipment must be sized based on the scrap density

available (Fig. 5). There is no point in buying a 5-tonne

feeder based on 2 tonnes per cubic metre, if the only scrap

available is 1 tonne per cubic metre or less.

Fig. 5: Typical scrap densities

Fig. 6: A possible charging system configuration.

The degree to which a charging system (Fig. 6) can be

mechanised is endless and depends on the degree of

automation required. Software to automate the system

is available from a number of suppliers (Fig.7). Different

materials can be stored in separate weigh hoppers. The

analysis of the various materials in the weigh hoppers can

be entered in the computer along with the formula of the

| 4 |


Fig. 7: Charge system computer screen.

grades of iron to be produced and the batch size. The

software will select a charge-based on the materials

available to ensure one get the maximum use of elements

already in the charge. Based on the feedback from a

spectrograph and the furnace load cells, the computer will

calculate the “trim” materials required. Once all the

necessary data has been loaded, a computer system will

generally provide the following information:

Weight and analysis of various materials in storage

weigh hoppers with maximum and minimum stock

levels.

Recovery rates for materials used.

Formulation for grades of iron to be produced.

The grade of iron currently being produced with

weights of the various charge materials required.

A report to a crane operator to indicate what

materials and weights are required.

The total weight of each charge.

Printed reports.

Weight of any trim material required.

The sequences in which multiple furnaces are to be

charged.

Information available from the computer system

can usually be accessed through the internet or a

LAN network.

When the weighed charge has been loaded on to the

furnace feeder, the feeder will index to a position behind

the furnace to be charged.

The only functions not automated are, indexing the feeder

forward over the furnace and starting the feeder. This is

for safety reasons to ensure the feeder is in the correct

position and the area is clear. As the melt progresses,

the operator needs to stop, and start the feeder to ensure

the furnace is always full but not overfilled.

Fig. 8: Melting at full power.

THE MELT CYCLE - the “melt cycle” starts as soon as

sufficient charge material is in the furnace to enable power

to be applied (Fig. 8). Modern power supplies

automatically control the power being applied to the

furnace as the condition in the furnace changes (Fig. 9).

The goal is to get the energy into the charge as quickly

and efficiently as possible. A power supply able to deliver

maximum power throughout the “melt cycle”, always

achieves the best melt rate. As the charge goes through

“Currie”, the voltage applied to the coil is allowed to

increase. This increase gives two advantages:

Fig. 9: Factors that will influence power usage.

Firstly, it ensures maximum kilowatts are continuously

applied to the coil. Secondly, a high coil voltage means

that the voltage induced into the charge is higher and

| 5 |


therefore the contact heating in the charge is more

efficient. Typically, this results in a ten percent

improvement in the melting rate as compared to a power

supply where the power draw drops as the charge passes

through “Currie”.

Fig.10: Melt cycle screen

Computer systems are available which will help manage

the “melt cycle” (Fig. 10). The final pouring temperature

is entered into the display. Taking a signal from the

furnace load cells and the kilowatts being applied to the

furnace, the computer’s algorithm displays a calculated

temperature as a vertical bar chart, which has set points

for various alarms and functions. As the display

temperature increases the operator is prompted to add

charge to the furnace. A temperature alarm is set at the

temperature the charge will be fully molten. This alarm

calls for a temperature dip. However, it is also the point

at which power should be turned off and the slag removed.

The computer system generally provides the following

functions and reports:

Manages an automated “melt cycle” (Fig. 10).

Starts the equipment at a predetermined time to

preheat the charge.

Provides an automatic sinter cycle (Fig. 11).

Provides diagnostics.

Displays a simulated bath temperature.

Displays information about the power supply.

Displays the weight of charge in the furnace.

Displays drain temperatures and alarms.

Maintains a record of alarm trips.

Provides printed reports.

Communicates via the internet or a LAN network.

Fig. 11: Sinter cycle screen

SLAG REMOVAL - During the removal of any slag, the

power must be off to ensure all the slag floats to the

surface and can be removed. The longer the power is off

the greater the effect on the overall “Utilisation”.

Manual removal can be enhanced by using a pearlite-

based slag coagulant, which will exfoliate to tie the slag

pieces together so they can be lifted off. Howeve,r it is a

hard and unpleasant job.

Fig. 12: Back slag removal.

If the furnace is larger, a back slag (Fig. 12) feature may

be available. In this case, the furnace is constructed with

| 6 |


two stanchions, which allow it to tilt in two directions,

and there is a second spout at the back opposite to the

pouring spout. When required, the furnace is back tilted

and the slag is raked into a slag bugging positioned under

the rear spout. The advantage of this system is that the

operator has to rake rather than lift, and the process can

be completed much quicker but there is a cost involved

at the time of purchase.

Another option is to use a slag grab (Fig. 13). This is a

pneumatically powered unit, which is suspended from a

crane or hoist above the furnace. It comprises of two

replaceable steel plate jaws that are opened and closed

by a pneumatic cylinder. With the jaws open the unit is

lowered into the slag, the jaws are closed and lifted out of

the furnace. It is then swung to the side and the jaws are

opened to drop the slag.

Fig. 13: A slag grab.

Fig.14: A mechanized slag grab.

A variation of the slag grab is a powered unit, which is

mounted on a structural steel frame that forms part of

the melter’s cabin between the furnaces (Fig. 14). The

unit can be swung to either side of the cabin to service

two furnaces. The jaw assembly is moved up and down

by a power cylinder and the action of the jaws is the same

as the crane mounted unit. This type of unit is useful if

the slag has a strong crust that needs a force to break

through the slag surface.

The key is to remove the slag as quickly and efficiently as

possible.

Fig. 15: Carbon, silicon and CE unit

TEMPERATURE DIP AND SAMPLE ANALYSIS -

in most iron foundries, a thermal analysis system is used

and the metal is poured based on the resulting carbon

equivalent reading and a carbon percentage (Fig. 15). If

a spectrograph analysis result is required then there will

be a delay, which adversely affects the untilisation.

Just after the temperature dip (Fig. 16) and analysis

sample are taken, holding power is restored to the

furnace.

Fig. 16: Dip temperature unit.

CHEMISTRY ADJUSTMENT AND SUPERHE-

ATING - depending on which analysis system is used,

| 7 |


just as the result is known, the “trim” materials should

be weighed and charged into the furnace.

If a spectrograph is to be used, the positioning of the unit

and the means by which a sample is delivered to the

laboratory and the result transmitted to the furnace

operator needs to be carefully thought out to ensure it

can be accomplished in the shortest possible time.

The carburiser used to “trim” needs to be small-grained

(Fig. 17) to increase its surface area and full-graphitised

as this ensures that it goes into solution quickly. The

function of carbon entering the iron matrix requires time,

temperature and a large contact surface. If a large-grained

petroleum coke is used at this point it will take longer to

enter the matrix and adversely affect the “utilisation”.

Silicon additions should be made using a crushed 75% Si

material.

Fig. 17 : Correctly sized “trim” graphite.

Immediately after the “trim” materials are in the furnace

full power should be able to superheat the charge and

create surface movement to increase the contact area. The

rate at which a given furnace capacity and power supply

will always superheat can be established, during the

commissioning. When the required degree of superheat

has been established, maximum power can be applied for

the required time and further temperature dip is not

necessary.

If the “melt cycle” is being managed by a computer a

mouse click will restart the power supply and the

computer will control the power to ensure only the

required amount of energy has been applied. At the

conclusion, an alarm will signal to alert the operator and

the power will be reduced to holding power.

ROBOTICS - software which allow a robot to carry out

a number of the these functions include:

Slag removal by raking or slag grab (Fig. 18)

·Temperature dip

Taking a liquid sample and pouring it into a thermal

analysis cup or chill mould

Adding pre weighed “trim” materials.

Fig. 18: A robot being used to remove slag.

The robot’s advantage is that it cannot be hurt, it does

not take rest breaks or holidays and it performs the same

way every time.

POURING - at the conclusion of the superheating/trim

addition cycle the furnace is ready to be poured empty.

If it is wanted to get the maximum advantage of the

contact heating as the charge goes through Currie one

must pour the furnace empty.

Very simply, the faster the furnace is emptied the better

(Fig. 19). During the pouring cycle, only low power is

applied and the time it takes to empty the furnace has

the greatest effect on the “utilisation”.

SUMMARY – If it is desired to purchase a specific

amount of energy from the utility supplier and biggest

profit is planned.

The one factor that will have the most effect is the level

of “utilisation” that can be achieved. Higher “utilisation”

| 8 |


is better. High utilisation means energy efficient

production cycles.

Fig. 20: A simple Power-Trak circuit.

EXAMPLE 1 - A 12,000 kW Power-Trak working with

a 17.5-tonne furnace with a hot lining and an empty

furnace will melt iron to 1,550oC at a rate of 25.2 tonnes

per hour at 100% “utilisation” +/-5% (Figs. 20 & 21).

It will melt 17.5 tonnes to 1,450oC in 39 minutes +/-5%.

A 12,000 kW will superheat 17.5 tonnes of iron through

100oC in 3 minutes +/-5%.

The 17.5-tonne furnace with a silica lining holding iron

at 1,550oC will require 250 kW per hour to overcome the

thermal losses +/-5%.

The following are two charts listing the operations

required for a full melt cycle (Fig. 22). In both cases, the

kilowatts and furnace size is the same. The time required

to melt and to superheat is the same in both.

Fig. 19 : A large capacity crane mounted ladle. Fig. 21: A typical melt and pour cycle for a Power-Trak

delivering batches to the casting plant.

Fig. 22 : Utilisation.

| 9 |


However, in the right-hand chart, (Fig.22) the time taken

for various operations is longer.

The calculation at the bottom shows the effect the

“utilisation” has on the production rate.

In both cases, one has to purchase a 12,000 kW supply

from the utility company. Yet only 40 - 50% of the total

production capacity can be achieved.

Fig. 23: A simple Dual-Trak circuit.

EXAMPLE 2 - A 12,500 kW Dual-Trak Plus has two

12,000 kW inverters (Fig. 23) working with two 17.5-

tonne furnaces and each of these will provide the same

performance as a single 12,000 kW Power-Trak. The extra

Fig. 24: Typical melt and pour sequence for a Dual-Trak delivering a

constant supply of metal to the casting plant.

500 kW in the AC/DC sections allows the unit to

simultaneously provide upto 500 kW to a second furnace

to maintain temperature.

Each furnace will melt and pour in sequence (Fig. 24). If

the “melt cycle” and the “non productive cycle” are equal

in duration, then you will achieve close to 100%

utilisation.

In practice, the utilisation of a Dual-Trak © power supply

is between 85 and 95% and there is a constant supply of

metal to the foundry casting process.

EXAMPLE 3 -A Tri-Trak has three inverters each rated

at 100% kW and three furnaces (Figs. 25 & Fig. 26). The

AC/DC section is rated at 210% of an inverter.

Each furnace will melt and pour in sequence. In this

case, to achieve 100% utilisation the “non-productive

cycle” must not exceed 50% of the “melt cycle” (Fig. 27).

THE MESSAGE: There is technology out there in

practice, materials, hardware and software that will

enable to run the equipment more efficiently. The more

efficiently one runs the equipment the higher will be the

| 10 |


“utilisation” of the equipment, and the KVA power supply

purchased from the utility company.

However, to achieve this, one will need properly trained

personnel and a management structure that ensures

training is kept upto date and the equipment is properly

and timely maintained.

Fig. 25: A simple Tri-Trak circuit.

Fig. 26 : Tri-Trak installation.

Fig. 27 : A typical melt and pour sequence for a Tri-Trak

delivering constant supply of metal to the casting plant.

technology improving your meltshop performancefoundryinfo-india.org/images/pdf/3b1.pdf · the...

Documents